" A 316 Stainless Steel ?piece of metal works well."
That's one of the things on my bucket ;list too-
Charles, it's amazing that you took to the new Si-PIN so well and right away.
By the way the SDD detector is in an on line now. Looks nice.Have only the one, no spare yet, too expensive!
Also found some premade cables for the Amptek PX-2 power supply that is hopefully going ton power the 2nd CdTe head. Lemo cables are hard to find in the right configuration.... we're waiting to see what housing the other fellows are going to use with the Si-PIN OEM kits before we try to standardize connectors or cases. The DP5 stack would best be served in a Hammond cast aluminum box milled out to allow the connectors to protrude, maybe someone with a home made CNC mill will help out....
Geo
----- Original Message -----
From: Dude <dfemer@...>
To:
[email protected]Cc: 'Mike Loughlin' <loughlin3@...>
Sent: Tue, 07 Jan 2020 23:43:08 -0500 (EST)
Subject: Re: [XRF] Tritium with Amptek Si-PIN vs NaI
Charles,
You missed the point, there is no doubt the ?SI Pin spectra are Zn
characteristic x-rays based on a good calibration and are well in spec.? The
NaI spectra aren¡¯t even calibrated in the composite spectra. It would be really
nice that when you are talking about a specific peak that you put the cursor on
it and leave out all the other spectra so we can actually determine energy and what
the data mean.?
You are convinced that its seeing the Zn and that¡¯s not saying it
isn¡¯t either. ?A scientific approach will prove the point not a personal belief
that you¡¯re going to base the rest of you calibrations on.? You have a
hypothesis -prove it. You¡¯ve got two signals overlapping each other in a
detector with very poor resolution.? You need to separate out the interferences
so you can determine definitively what you are looking at and determine the best
calibration point. Using the NaI detector run a pure Zn target with a non brem
source (AM-241 would be a good source) then with the Tritium source. Take both
spectra sequentially to avoid E drift.? Overlay it over the Tritium ?Brem
spectra. ?Interpret that data and come to a conclusion independent of your bias
A Fe-55 check source would be useful to bracket the Am-241 energy
cal in order to get a tight E-cal.? It ?would be even more useful for the low energy
cal of the NaI detector rather than a weak tritium light.
You always need to do a Energy Cal check every time you use the
instrument and even at the end of the day - or at some point it¡¯s going to bite
you.? A 316 Stainless Steel ?piece of metal works well.
Dud
?
?
Attached is the same tritium scan with Cu and Au
characteristic x-rays bracketing the Zn peaks.? If any doubt remains that
those are Zn peaks here is a high resolution scan I have seen around the
internet. Also, I don't really care about the shape of the brem.? I am
convinced that what I am seeing with my NaI is actually the Zn peaks merged
together at about 9keV.? I think that is a better estimate than the 11keV I
was using before.
BTW, I have really not needed to calibrate this Si-PIN at
all.? Geo sent it to me fully calibrated.? So far everything I have
tried checks out.
?
On Tue, Jan 7, 2020 at 4:50 PM Dude <dfemer@...> wrote:
Charles,
You may
be close to what that cal peak will be but it isn¡¯t necessarily accurate. The
shape of the brem is going to be dependent on the filtering its undergone from
both the glass and the different detector¡¯s windows.? To do a proper cal
use a monochromatic source of the proper energy for each detector. Getting the
cal points right saves a lot of interpretation grief later, especially with
high resolution detectors.? Spend the money and do it right they say. Get
a Fe 55 disk or light up a characteristic x-ray
Dud
?
?
I
have been playing with the Si-PIN that I got from Geo and am getting some great
results.? When measuring a tritium keychain with NaI I get a strong peak
down in the teens (dark blue) but I have never been sure where to place that
peak given that this is a broad Bremsstrahlung region.? I had been placing
it at 11keV because this is supposed to be in the middle of that region.
However,
my Si-PIN results show how dominant the Zn sulphide peaks are so now I would
place my NaI peak right between them at 9keV.? That makes for a great
calibration point at the extreme low end of my NaI setup.
